Inhibition Equivalency Factors for Dinophysistoxin ... - ACS Publications

Feb 13, 2013 - In this work, the applicability of two PPIAs to the determination of okadaic acid (OA), dinophysistoxin-1 (DTX-1), dinophysistoxin-2 (D...
0 downloads 0 Views 431KB Size
Article pubs.acs.org/JAFC

Inhibition Equivalency Factors for Dinophysistoxin‑1 and Dinophysistoxin‑2 in Protein Phosphatase Assays: Applicability to the Analysis of Shellfish Samples and Comparison with LC-MS/MS Diana Garibo, Pablo de la Iglesia, Jorge Diogène, and Mònica Campàs* IRTA, Carretera de Poble Nou, km 5.5, 43540 Sant Carles de la Ràpita, Spain S Supporting Information *

ABSTRACT: The protein phosphatase inhibition assay (PPIA) is a well-known strategy for the determination of diarrheic shellfish poisoning (DSP) lipophilic toxins, which deserves better characterization and understanding to be used as a routine screening tool in monitoring programs. In this work, the applicability of two PPIAs to the determination of okadaic acid (OA), dinophysistoxin-1 (DTX-1), dinophysistoxin-2 (DTX-2), and their acyl ester derivatives in shellfish has been investigated. The inhibitory potencies of the DSP toxins on a recombinant and a wild PP2A have been determined, allowing the establishment of inhibition equivalency factors (IEFs) (1.1 and 0.9 for DTX-1, and 0.4 and 0.6 for DTX-2, for recombinant and wild PP2A, respectively). The PPIAs have been applied to the determination of OA equivalent contents in spiked and naturally contaminated shellfish samples. Results have been compared to those obtained by LC-MS/MS analysis, after application of the IEFs, showing good agreements. KEYWORDS: protein phosphatase 2A, protein phosphatase inhibition assay, okadaic acid, dinophysistoxin-1, dinophysistoxin-2, liquid chromatography−tandem mass spectrometry



contents in shellfish.9 This regulation was applied from July 1, 2011, and the LC-MS/MS method will replace the MBA in 2015. These regulations also accept the use of other chemical methods, as well as immunoassays and functional assays, such as the protein phosphatase inhibition assay (PPIA), as alternatives or supplements to the LC-MS/MS method, provided that they can determine OA, DTX-1, DTX-2, and their esters, that they fulfill the method performance criteria (they should be validated and successfully tested under a recognized proficiency test scheme), and that their implementation provides an equivalent level of public health protection. The development of rapid, sensitive, and low-cost methods for the detection of DSP toxins is necessary to guarantee shellfish safety and protect human health. The PPIA is an interesting method for the simple, fast, sensitive, and robust determination of DSP toxin contents in shellfish. Colorimetric PPIAs using PP in solution have been developed.10−17 In most of the works, OA has been used as model DSP toxin,10,11 and only in a few studies has the inhibitory potential of OA analogues or derivatives been evaluated and none of them have used high-quality certified reference materials.12−17 The establishment and use of toxicity equivalent factors (TEFs) for toxic compounds of a same group in alternative methods for marine toxin detection is necessary to guarantee consumer protection in monitoring programs, as they allow a better estimation of the toxic potential of a mixture of toxins with different potency.18,19

INTRODUCTION Okadaic acid (OA) and its analogues dinophysistoxin-1 (DTX1) and dinophysistoxin-2 (DTX-2) are lipophilic phycotoxins produced mainly by dinoflagellates of the genera Dinophysis and Prorocentrum.1 Their chemical structure is composed of a polyketide backbone containing furan and pyran-type ether rings and an α-hydroxycarboxyl function, the difference between analogues being only the number or the position of the methyl groups.2 When incorporated in shellfish, these phycotoxins are accumulated mainly in the digestive gland and are responsible for the diarrheic shellfish poisoning (DSP) syndrome, which causes gastrointestinal disturbances such as diarrhea, nausea, vomiting, and abdominal pain.3 OA and DTXs are known inhibitors of protein phosphatases (PP1 and PP2A), enzymes that play an important role in protein dephosphorylation in cells.4 These toxins bind to the receptor site of the enzyme, blocking its activity, and as a consequence they favor hyperphosphorylation of proteins that control sodium secretion and of cytoskeletal or junctional moieties that regulate solute permeability, causing sodium release and a subsequent passive loss of fluids responsible for the diarrheic symptoms.5 Moreover, it has been demonstrated to be an additional tumor promoter in mouse skin carcinogenesis.6 Because of their implications on public health, the Regulation (EC) No. 853/2004 in Europe has established a maximum permitted level of 160 μg of OA equivalents/kg shellfish meat.7 Although it is possible to use the mouse bioassay (MBA)8 until December 31, 2014, the Commission Regulation (EU) No. 15/ 2011 has recently established that liquid chromatography− tandem mass spectrometry (LC-MS/MS) should be applied as the reference method for the determination of lipophilic toxin © 2013 American Chemical Society

Received: Revised: Accepted: Published: 2572

December 13, 2012 February 4, 2013 February 13, 2013 February 13, 2013 dx.doi.org/10.1021/jf305334n | J. Agric. Food Chem. 2013, 61, 2572−2579

Journal of Agricultural and Food Chemistry

Article

for 3 min at maximum speed level (∼2500 laps/min). Extract was centrifuged at 2000g for 10 min at ca. 20 °C, and the supernatant was transferred to a 20-mL volumetric flask. The extraction of the residual tissue pellet was repeated with 100% MeOH (9 mL), and the sample was homogenized for 1 min with a high-speed homogenizer Ultraturrax T25 (IKA-Labortechnik). After centrifugation under the same conditions previously applied, the supernatant was transferred and combined with the supernatant from the previous extraction, and the total extract was made up to 20 mL with 100% MeOH in a volumetric flask. Extracts were passed through 0.2 μm cutoff nylon syringe filters (Whatman), and were directly injected onto the LCMS/MS system. For extracts to be tested with PPIA, samples were evaporated in a Speed VAC concentrator (Organomation Associates, Inc., Berlin, MA) under nitrogen at room temperature, and the residues were resuspended in the corresponding buffer. Sample Hydrolysis. To determine the total OA and DTX content, an alkaline hydrolysis was performed before LC-MS/MS analysis and PPIA.20,21 For the hydrolysis, NaOH at 2.5 M (125 μL) was added to the extract (1.25 mL), the mixture was homogenized in a vortex mixer for 0.5 min, and heated at 76 °C for 40 min in a Multi-BlockHeater from Lab-line Instruments, Inc. (Maharashtra, India). After cooling to room temperature, HCl at 2.5 M (125 μL) was added for neutralization and the sample was homogenized by vortex-mixing for 0.5 min. The hydrolyzed extract was then filtered through 0.2 μm cutoff nylon syringe filters (Whatman). As described for crude extracts, hydrolyzed extracts were directly analyzed by LC-MS/MS, while for PPIA they were evaporated under nitrogen and resuspended in the corresponding buffer to the desired concentration. Colorimetric PPIA. The PPIA was performed as described previously17,22 but three different experiments were carried out: (1) evaluation of the inhibitory potencies of DSP toxins and the IEFs of DTXs, (2) determination of DSP toxin contents in spiked mussel samples, and (3) determination of DSP toxin contents in naturally contaminated shellfish samples. To this purpose, 50 μL of blank mussel sample solution at 12.5 mg/mL spiked with OA, DTX-1, or DTX-2 standard solutions at different concentrations ranging from 1.6 to 100.0 μg/L (for experiment 1 and for OA calibration curves of experiments 2 and 3), 50 μL of blank mussel sample solution at 12.5 mg/mL spiked with OA at 160 μg/L, DTX-1 at 166 μg/L and/or DTX-2 at 176 μg/L (for experiment 2) or 50 μL of naturally contaminated shellfish sample solution at different concentrations ranging from 1.6 to 12.5 mg/mL (for experiment 3) were added into microtiter wells containing 100 μL of PP2A solution (recombinant from GTP Technology or wild from Upstate Biotechnology) at 1.25 U/mL. Then, 50 μL of 25 mM p-NPP solution were added and after 1-h incubation at 22 °C in the dark, the absorbance at 405 nm was measured with an automated multiwell scanning spectrophotometer (Biotek, Synergy HT, Winooski, VT). Samples were prepared in a buffer solution containing 30 mM Tris-HCl, 20 mM MgCl2, pH 8.4. PP2A, and p-NPP solutions were prepared in the same buffer, also containing 2 mM DTT and 0.2 mg/mL BSA. Assays were performed in triplicate. In the analysis of naturally contaminated shellfish samples and in the evaluation of the inhibitory effect of DSP toxin mixtures, OA calibration curves using 12.5 and 6.3 mg/mL of mussel matrix, for recombinant and wild PP2A, respectively, were always performed in parallel for the precise toxin quantification. The DSP toxin calibration curves obtained by PPIA were analyzed with SigmaPlot software package 10.0 and fitted to sigmoidal logistic four-parameter equations: a y = y0 + 1 + (x /x0)b

Apart from the use of the PPIA for quantitative purposes, this assay is a promising screening tool to be run in parallel to the official control methods in monitoring programs. For example, PPIA could be used to screen DSP toxins in hydrolyzed shellfish samples, reducing the instrumental analytical requirements and still protecting public health. Nevertheless, it requires in-depth characterization and performance evaluation before its approval and routine use. With this aim, we have evaluated the practical application of the PPIA to the analysis of shellfish contaminated with DSP toxins. First, the different inhibitory potencies of OA, DTX-1, and DTX-2 on a recombinant and a wild PP2A have been determined, and the corresponding inhibition equivalency factors (IEFs) have been established. Definition of IEFs is important to characterize the performance of the PPIA, but it is also crucial to get comparable results with the reference LC-MS/MS method. Afterward, both PPIAs have been applied to the determination of DSP toxin contents in mussel samples spiked with OA, DTX-1, and/or DTX-2 and in naturally contaminated shellfish (mussels, cockles, clams, and razor clams) samples. Results have been compared with those obtained by LC-MS/MS analysis after the application of the IEFs.



MATERIALS AND METHODS

Reagents and Materials. Certified calibration solution of okadaic acid (NRC CRM-OA, 14 300 μg/L), dinophysistoxin-1 (NRC CRMDTX1, 15 100 μg/L), and dinophysistoxin-2 (NRC CRM-DTX2, 7800 μg/L) in methanol (MeOH) were kindly provided by the Institute for Marine Biosciences of the National Research Council (Halifax, Canada). The recombinant protein phosphatase 2A (PP2A) catalytic subunit was produced by Gene to Protein (GTP) Technology (Toulouse, France). Commercial protein phosphatase 2A (PP2A), isolated as the heterodimer of 60 kDa and 36 kDa subunits from human red blood cells, was obtained from Upstate Biotechnology (New York, NY). The activity of the stock solutions was between 766 and 1025 U/mL for PP2A from GTP Technology and between 5720 and 7491 U/mL for PP2A from Upstate Biotechnology, 1 U being defined as the amount of enzyme required to hydrolyze of 1 nmol pnitrophenyl phosphate (p-NPP) in 1 min at room temperature. Components of buffers and p-NPP were purchased from Sigma (Tres Cantos, España). For LC-MS/MS analysis, gradient-grade MeOH, formic acid, and hypergrade acetonitrile (ACN) for LC-MS were purchased from Merck (Darmstadt, Germany). Ammonium formate (≥99.995%), sodium hydroxide pellets (≥99%), and hydrochloric acid 37% for analysis were purchased from Sigma-Aldrich (St. Louis, MO), Riedel-de Haën (Seelze, Germany), and Panreac (Barcelona, Spain), respectively. All solutions were prepared using Milli-Q grade water obtained from a Millipore purification system (Bedford, MA). Samples. Fourteen samples obtained from the European Union Reference Laboratory for Marine Biotoxins (EU-RL-MB) in Vigo, Spain, and that had been analyzed for the collaborative Interlaboratory Validation Study of the “EU-Harmonised Standard Operation Procedure for Determination of Lipophilic Marine Biotoxins in Molluscs by LC-MS/MS” (EU-RL-MB SOP),20 were used in this work. They corresponded to seven materials distributed as blind duplicates of different species of molluscs naturally contaminated: raw wedge shell clam homogenate (Donax trunculus), raw razor clam homogenate (Ensis acuatus), raw mussel homogenate (Mytilus edulis), raw stripped venus (Chamelea gallina), two cooked mussel homogenates (Mytilus edulis), and raw cockle homogenate (Cerastoderma edule). Lipophilic Toxin Extraction. For lipophilic toxin extraction, the “EU-Harmonised Standard Operation Procedure for Determination of Lipophilic Marine Biotoxins in Molluscs by LC-MS/MS” (EU-RL-MB SOP) procedure was followed.20 First, tissue homogenate (2 g) was weighed in a 50-mL polypropylene centrifuge tube. MeOH (100%, 9 mL) was added, and the sample was homogenized by vortex-mixing

where a and y0 are the asymptotic maximum and minimum values, respectively, x0 is the x value at the inflection point, and b is the slope at the inflection point. LC-MS/MS Analysis. Liquid chromatography−tandem mass spectrometry (LC-MS/MS) was applied following the “EU-Harmonised Standard Operation Procedure for Determination of Lipophilic Marine Biotoxins in Molluscs by LC-MS/MS” (EU-RL-MB SOP).20 Analyses were conducted on an Agilent 1200 LC (Agilent Technologies, Santa Clara, CA) coupled with a 3200 QTRAP mass 2573

dx.doi.org/10.1021/jf305334n | J. Agric. Food Chem. 2013, 61, 2572−2579

Journal of Agricultural and Food Chemistry

Article

Figure 1. Dose−response curves for the inhibition of recombinant (A) and wild (B) PP2A by OA, DTX-1, and DTX-2. Inhibition is expressed as percentage of the control (no toxin); x values refer to initial toxin concentrations.

Table 1. Curve Parameters Derived from the Sigmoidal Logistic Four-Parameter Fitting for the Inhibition of PP2As by OA, DTX-1, and DTX-2 toxin OA DTX-1 DTX-2

enzyme recombinant wild recombinant wild recombinant wild

IC50 (μg/L) 2.93 1.54 2.90 1.66 7.54 3.38

IEF

working range IC20 − IC80 (μg/L)

1.0 1.0 1.1 0.9 0.4 0.6

1.2−8.4 0.7−3.5 1.4−6.0 0.9−2.8 2.2−29.7 1.1−9.2

spectrometer through a TurboV electrospray ion source (Applied Biosystems, Foster City, CA). Chromatographic separations were performed at 30 °C and 0.2 mL/min on a Luna C8(2) column (50 mm × 1 mm, 3 μm) protected with a SupelcoGuard C8(2) cartridge (4 mm × 2 mm, 3 μm), both from Phenomenex (Torrance, CA). Acidic chromatographic elution was selected with mobile phases 100% water (A) and 95% acetonitrile (B), both containing 2 mM ammonium formate and 50 mM formic acid. For DSP toxins, multiple reactions monitoring (MRM) analysis was accomplished from the precursor ions 803.5 and 817.5 m/z for OA/DTX-2 and DTX-1, respectively. Product ions were common for all DSP toxins, with ions 255.2 m/z monitored for quantification and 113.1 m/z acquired for confirmatory purposes. The mass spectrometer was operated in negative polarity, and compound-dependent parameters for MS/MS detection were tuned on the mass spectrometer through direct infusion of the CRM-OA standard: declustering potential −115 V, entrance potential −12 V (for 803.5 > 255.2) and −10.5 V (for 803.5 > 113.1), collision entrance potential collision energy −64 V (for 803.5 > 255.2) and −68 V (for 803.5 > 113.1), and collision cell exit potential −4 V. Gas/source parameters were also optimized (curtain gas 20 psi; ion spray −4500 V, temperature 400 °C, nebulizer gas 50 psi, heater gas 50 psi). Under these conditions, the LOD and LOQ were 10 and 30 μg/kg OA in shellfish meat, respectively. The quantification curve obtained for OA was used also for the quantification of DTX-1 and DTX-2, because this approach was the recommended by the EU-RL-MB SOP.20 Analyst software v1.4.2 was used for the entire MS tune, instrument control, data acquisition, and data analysis. Statistical Analysis. To evaluate differences in the calibration curves for OA, DTX-1, and DTX-2 between recombinant and wild PP2A, the paired t test was used (N = 10). Differences in the results were considered statistically significant at the 0.05 level. Prior to analysis, data were tested for normality; Wilcoxon matched-pairs signed-ranks test was used for non-normally distributed data sets instead of the paired t test.

R2

equation y y y y y y

= = = = = =

−1.5

−2.0 + (96.1/(1 + x/2.7) ) −0.7 + (98.4/(1 + x/1.5)−1.8) 0.3 + (95.5/(1 + x/2.7)−2.0) 0.9 + (96.7/(1 + x/1.6)−2.6) −2.1 + (92.8/(1 + x/5.6)−1.2) −1.3 + (98.4/(1 + x/2.9)−1.4)

0.9994 0.9998 0.9996 0.9996 0.9996 0.9996

To evaluate the correlation between the OA equivalent contents in spiked mussel samples determined from the two PPIAs and the expected values, the linear regression model was used. The linear regression model was also used to evaluate the correlation between the OA equivalent contents in naturally contaminated shellfish samples determined from the two PPIAs and the values obtained from the LCMS/MS analysis after application of the IEFs for each PP2A and the TEFs from EFSA. Differences in the results were considered statistically significant also at the 0.05 level. The SigmaStat software package 3.1 was used for the paired t tests, the Wilcoxon matched-pairs signed-ranks tests, and the linear regressions.



RESULTS AND DISCUSSION Inhibitory Potencies of DSP Toxins and IEFs of DTXs. Dose−response curves with OA, DTX-1, and DTX-2 (Figure 1) were performed to evaluate the inhibitory potencies of these DSP toxins on the activity of two PP2A enzymes from different origins (recombinant and wild). Toxin dilutions from stock solutions were prepared in a buffer solution containing blank mussel matrix at 12.5 mg/mL. This blank mussel did not contain DSP lipophilic toxins as determined by LC-MS/MS. The 12.5 mg/mL concentration had been previously established as equal to (for wild PP2A) or below (for recombinant PP2A) the maximum loading limit to use in the PPIA to avoid unspecific inhibition from the mussel matrix.22 In Table 1, the 50% inhibition coefficient (IC50) values, the inhibition equivalency factors (IEFs), and the working ranges (defined between IC20 and IC80) are presented together with the equations and the corresponding R2 values. The IEFs were calculated as the ratios of the IC50 for OA to the IC50 for DTX1 or DTX-2, for each enzyme. Comparing enzyme sources, the wild PP2A was significantly more sensitive to all DSP toxins than the recombinant one (tOA 2574

dx.doi.org/10.1021/jf305334n | J. Agric. Food Chem. 2013, 61, 2572−2579

Journal of Agricultural and Food Chemistry

Article

Table 2. DSP Lipophilic Toxin Spiking Combinations, OA Equivalent Contents (μg of OA equiv/kg mussel meat) Expected According to the Spiked Concentrations and the IEFs and Determined by PPIA with Recombinant and Wild PP2Aa DSP lipophilic toxin

expected [OA]eq

determined [OA]eq

combination

OA

DTX-1

DTX-2

PP2Arec

PP2Awild

PP2Arec

PP2Awild

1 2 3 4 5 6 7 8

+ + + − + − − −

+ + − + − + − −

+ − + + − − + −

404 340 224 244 160 180 64 0

404 307 257 244 160 147 97 0

576 ± 3 314 ± 16 277 ± 18 187 ± 9 172 ± 3 101 ± 6 n.d.b n.d.

569 ± 17 328 ± 3 273 ± 1 208 ± 4 164 ± 1 127 ± 9 63 ± 2 n.d.

a The + symbol indicates 160, 166, and 176 μg/kg for OA, DTX-1, and DTX-2, respectively; the − symbol indicates 0. bn.d. = not detected: